External Memory Interfaces Cyclone® 10 GX FPGA IP User Guide

Download PDF
ID 683663
Date 4/01/2024
Public
Document Table of Contents
Document Table of Contents x
1. Release Information 2. External Memory Interfaces Cyclone® 10 GX FPGA IP Introduction 3. Cyclone® 10 GX EMIF IP Product Architecture 4. Cyclone® 10 GX EMIF IP End-User Signals 5. Cyclone® 10 GX EMIF – Simulating Memory IP 6. Cyclone® 10 GX EMIF IP for DDR3 7. Cyclone® 10 GX EMIF IP for LPDDR3 8. Cyclone® 10 GX EMIF IP Timing Closure 9. Optimizing Controller Performance 10. Cyclone® 10 GX EMIF IP Debugging 11. External Memory Interfaces Cyclone® 10 GX FPGA IP User Guide Archives 12. Document Revision History for Cyclone® 10 GX External Memory Interfaces FPGA IP User Guide
2. External Memory Interfaces Cyclone® 10 GX FPGA IP Introduction x
2.1. Cyclone® 10 GX EMIF IP Design Flow 2.2. Cyclone® 10 GX EMIF IP Design Checklist
3. Cyclone® 10 GX EMIF IP Product Architecture x
3.1. EMIF Architecture: Introduction 3.2. Cyclone® 10 GX EMIF Sequencer 3.3. Cyclone® 10 GX EMIF Calibration 3.4. Cyclone® 10 GX EMIF Controller 3.5. Hardware Resource Sharing Among Multiple EMIFs 3.6. Cyclone® 10 GX EMIF Ping Pong PHY
3.1. EMIF Architecture: Introduction x
3.1.1. I/O Subsystem 3.1.2. I/O Column 3.1.3. I/O AUX 3.1.4. I/O Bank 3.1.5. I/O Lane 3.1.6. Input DQS Clock Tree 3.1.7. PHY Clock Tree 3.1.8. PLL Reference Clock Networks 3.1.9. Clock Phase Alignment
3.1.4. I/O Bank x
3.1.4.1. Implementing a x8 Interface with Hard Memory Controller 3.1.4.2. Implementing a x72 Interface with Hard Memory Controller
3.2. Cyclone® 10 GX EMIF Sequencer x
3.2.1. DQS Tracking
3.3. Cyclone® 10 GX EMIF Calibration x
3.3.1. Calibration Stages 3.3.2. Calibration Stages Descriptions 3.3.3. Calibration Algorithms 3.3.4. Calibration Flowchart
3.4. Cyclone® 10 GX EMIF Controller x
3.4.1. Hard Memory Controller 3.4.2. Hard Memory Controller Rate Conversion Feature
3.4.1. Hard Memory Controller x
3.4.1.1. Hard Memory Controller Features 3.4.1.2. Hard Memory Controller Main Control Path 3.4.1.3. Data Buffer Controller
3.5. Hardware Resource Sharing Among Multiple EMIFs x
3.5.1. I/O Aux Sharing 3.5.2. I/O Bank Sharing 3.5.3. PLL Reference Clock Sharing 3.5.4. Core Clock Network Sharing
3.6. Cyclone® 10 GX EMIF Ping Pong PHY x
3.6.1. Ping Pong PHY Feature Description 3.6.2. Ping Pong PHY Architecture 3.6.3. Ping Pong PHY Limitations 3.6.4. Ping Pong PHY Calibration 3.6.5. Using the Ping Pong PHY 3.6.6. Ping Pong PHY Simulation Example Design
4. Cyclone® 10 GX EMIF IP End-User Signals x
4.1. Interface and Signal Descriptions 4.2. AFI Signals 4.3. AFI 4.0 Timing Diagrams 4.4. Cyclone® 10 GX Memory Mapped Register (MMR) Tables
4.1. Interface and Signal Descriptions x
4.1.1. Intel Cyclone 10 GX EMIF IP Interfaces for DDR3 4.1.2. Intel Cyclone 10 GX EMIF IP Interfaces for LPDDR3
4.1.1. Intel Cyclone 10 GX EMIF IP Interfaces for DDR3 x
4.1.1.1. pll_ref_clk for DDR3 4.1.1.2. pll_locked for DDR3 4.1.1.3. pll_extra_clk_0 for DDR3 4.1.1.4. pll_extra_clk_1 for DDR3 4.1.1.5. pll_extra_clk_2 for DDR3 4.1.1.6. pll_extra_clk_3 for DDR3 4.1.1.7. oct for DDR3 4.1.1.8. mem for DDR3 4.1.1.9. status for DDR3 4.1.1.10. afi_reset_n for DDR3 4.1.1.11. afi_clk for DDR3 4.1.1.12. afi_half_clk for DDR3 4.1.1.13. afi for DDR3 4.1.1.14. emif_usr_reset_n for DDR3 4.1.1.15. emif_usr_clk for DDR3 4.1.1.16. emif_usr_reset_n_sec for DDR3 4.1.1.17. emif_usr_clk_sec for DDR3 4.1.1.18. cal_debug_reset_n for DDR3 4.1.1.19. cal_debug_clk for DDR3 4.1.1.20. cal_debug_out_reset_n for DDR3 4.1.1.21. cal_debug_out_clk for DDR3 4.1.1.22. clks_sharing_master_out for DDR3 4.1.1.23. clks_sharing_slave_in for DDR3 4.1.1.24. clks_sharing_slave_out for DDR3 4.1.1.25. ctrl_amm for DDR3 4.1.1.26. ctrl_auto_precharge for DDR3 4.1.1.27. ctrl_user_priority for DDR3 4.1.1.28. ctrl_ecc_user_interrupt for DDR3 4.1.1.29. ctrl_ecc_readdataerror for DDR3 4.1.1.30. ctrl_mmr_slave for DDR3 4.1.1.31. cal_debug for DDR3 4.1.1.32. cal_debug_out for DDR3
4.1.2. Intel Cyclone 10 GX EMIF IP Interfaces for LPDDR3 x
4.1.2.1. pll_ref_clk for LPDDR3 4.1.2.2. pll_locked for LPDDR3 4.1.2.3. pll_extra_clk_0 for LPDDR3 4.1.2.4. pll_extra_clk_1 for LPDDR3 4.1.2.5. pll_extra_clk_2 for LPDDR3 4.1.2.6. pll_extra_clk_3 for LPDDR3 4.1.2.7. oct for LPDDR3 4.1.2.8. mem for LPDDR3 4.1.2.9. status for LPDDR3 4.1.2.10. afi_reset_n for LPDDR3 4.1.2.11. afi_clk for LPDDR3 4.1.2.12. afi_half_clk for LPDDR3 4.1.2.13. afi for LPDDR3 4.1.2.14. emif_usr_reset_n for LPDDR3 4.1.2.15. emif_usr_clk for LPDDR3 4.1.2.16. cal_debug_reset_n for LPDDR3 4.1.2.17. cal_debug_clk for LPDDR3 4.1.2.18. cal_debug_out_reset_n for LPDDR3 4.1.2.19. cal_debug_out_clk for LPDDR3 4.1.2.20. clks_sharing_master_out for LPDDR3 4.1.2.21. clks_sharing_slave_in for LPDDR3 4.1.2.22. clks_sharing_slave_out for LPDDR3 4.1.2.23. ctrl_user_priority for LPDDR3 4.1.2.24. ctrl_mmr_slave for LPDDR3 4.1.2.25. cal_debug for LPDDR3 4.1.2.26. cal_debug_out for LPDDR3
4.2. AFI Signals x
4.2.1. AFI Clock and Reset Signals 4.2.2. AFI Address and Command Signals 4.2.3. AFI Write Data Signals 4.2.4. AFI Read Data Signals 4.2.5. AFI Calibration Status Signals 4.2.6. AFI Shadow Register Management Signals
4.3. AFI 4.0 Timing Diagrams x
4.3.1. AFI Address and Command Timing Diagrams 4.3.2. AFI Write Sequence Timing Diagrams 4.3.3. AFI Read Sequence Timing Diagrams 4.3.4. AFI Calibration Status Timing Diagram
4.4. Cyclone® 10 GX Memory Mapped Register (MMR) Tables x
4.4.1. ctrlcfg0 4.4.2. ctrlcfg1 4.4.3. dramtiming0 4.4.4. sbcfg1 4.4.5. caltiming0 4.4.6. caltiming1 4.4.7. caltiming2 4.4.8. caltiming3 4.4.9. caltiming4 4.4.10. caltiming9 4.4.11. dramaddrw 4.4.12. sideband0 4.4.13. sideband1 4.4.14. sideband2 4.4.15. sideband3 4.4.16. sideband4 4.4.17. sideband5 4.4.18. sideband6 4.4.19. sideband7 4.4.20. sideband8 4.4.21. sideband9 4.4.22. sideband10 4.4.23. sideband11 4.4.24. sideband12 4.4.25. sideband13 4.4.26. dramsts 4.4.27. niosreserve0 4.4.28. niosreserve1 4.4.29. ecc3: ECC Error and Interrupt Configuration 4.4.30. ecc4: Status and Error Information 4.4.31. ecc5: Address of Most Recent SBE/DBE 4.4.32. ecc6: Address of Most Recent Correction Command Dropped
5. Cyclone® 10 GX EMIF – Simulating Memory IP x
5.1. Simulation Options 5.2. Simulation Walkthrough
5.2. Simulation Walkthrough x
5.2.1. Calibration Modes 5.2.2. Abstract PHY Simulation 5.2.3. Simulation Scripts 5.2.4. Functional Simulation with Verilog HDL 5.2.5. Functional Simulation with VHDL 5.2.6. Simulating the Design Example
6. Cyclone® 10 GX EMIF IP for DDR3 x
6.1. Parameter Descriptions 6.2. Board Skew Equations 6.3. Pin and Resource Planning 6.4. DDR3 Board Design Guidelines
6.1. Parameter Descriptions x
6.1.1. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: General 6.1.2. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Memory 6.1.3. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Mem I/O 6.1.4. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: FPGA I/O 6.1.5. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Mem Timing 6.1.6. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Board 6.1.7. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Controller 6.1.8. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Diagnostics 6.1.9. Intel Cyclone 10 GX EMIF IP DDR3 Parameters: Example Designs
6.2. Board Skew Equations x
6.2.1. Equations for DDR3 Board Skew Parameters
6.3. Pin and Resource Planning x
6.3.1. Interface Pins 6.3.2. FPGA Resources 6.3.3. Pin Guidelines for Cyclone® 10 GX EMIF IP
6.3.1. Interface Pins x
6.3.1.1. Estimating Pin Requirements 6.3.1.2. DIMM Options 6.3.1.3. Maximum Number of Interfaces
6.3.2. FPGA Resources x
6.3.2.1. OCT 6.3.2.2. PLL
6.3.3. Pin Guidelines for Cyclone® 10 GX EMIF IP x
6.3.3.1. General Guidelines 6.3.3.2. x4 DIMM Implementation 6.3.3.3. Command and Address Signals 6.3.3.4. Clock Signals 6.3.3.5. Data, Data Strobes, DM/DBI, and Optional ECC Signals 6.3.3.6. Resource Sharing Guidelines (Multiple Interfaces) 6.3.3.7. Ping-Pong PHY Implementation
6.4. DDR3 Board Design Guidelines x
6.4.1. Terminations and Slew Rates with Cyclone® 10 GX Devices 6.4.2. Channel Signal Integrity Measurement 6.4.3. Layout Approach 6.4.4. Design Layout Guidelines 6.4.5. Package Deskew
6.4.1. Terminations and Slew Rates with Cyclone® 10 GX Devices x
6.4.1.1. Dynamic On-Chip Termination (OCT) in Cyclone® 10 Devices 6.4.1.2. Choosing Terminations on Cyclone® 10 Devices 6.4.1.3. On-Chip Termination Recommendations for Cyclone® 10 Devices 6.4.1.4. Slew Rates
6.4.2. Channel Signal Integrity Measurement x
6.4.2.1. Importance of Accurate Channel Signal Integrity Information 6.4.2.2. Understanding Channel Signal Integrity Measurement 6.4.2.3. How to Enter Calculated Channel Signal Integrity Values 6.4.2.4. Guidelines for Calculating DDR3 Channel Signal Integrity
6.4.4. Design Layout Guidelines x
6.4.4.1. General Layout Guidelines 6.4.4.2. Layout Guidelines 6.4.4.3. Length Matching Rules 6.4.4.4. Spacing Guidelines 6.4.4.5. Layout Guidelines for DDR3 SDRAM Wide Interface (>72 bits) 6.4.4.6. Fly-By Network Design for Clock, Command, and Address Signals
6.4.5. Package Deskew x
6.4.5.1. DQ/DQS/DM Deskew 6.4.5.2. Address and Command Deskew 6.4.5.3. Package Deskew Recommendations for Cyclone® 10 Devices 6.4.5.4. Deskew Example 6.4.5.5. Package Migration
7. Cyclone® 10 GX EMIF IP for LPDDR3 x
7.1. Parameter Descriptions 7.2. Board Skew Equations 7.3. Pin and Resource Planning 7.4. LPDDR3 Board Design Guidelines
7.1. Parameter Descriptions x
7.1.1. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: General 7.1.2. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Memory 7.1.3. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Mem I/O 7.1.4. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: FPGA I/O 7.1.5. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Mem Timing 7.1.6. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Board 7.1.7. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Controller 7.1.8. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Diagnostics 7.1.9. Intel Cyclone 10 GX EMIF IP LPDDR3 Parameters: Example Designs
7.2. Board Skew Equations x
7.2.1. Equations for LPDDR3 Board Skew Parameters
7.3. Pin and Resource Planning x
7.3.1. Interface Pins
7.3.1. Interface Pins x
7.3.1.1. Estimating Pin Requirements 7.3.1.2. Maximum Number of Interfaces 7.3.1.3. FPGA Resources 7.3.1.4. OCT 7.3.1.5. PLL 7.3.1.6. Pin Guidelines for Cyclone® 10 GX EMIF IP
7.3.1.6. Pin Guidelines for Cyclone® 10 GX EMIF IP x
7.3.1.6.1. General Guidelines 7.3.1.6.2. LPDDR3 Clock Signal 7.3.1.6.3. LPDDR3 Command and Address Signal 7.3.1.6.4. LPDDR3 Data, Data Strobe, and DM Signals 7.3.1.6.5. Resource Sharing Guidelines (Multiple Interfaces)
7.4. LPDDR3 Board Design Guidelines x
7.4.1. Terminations and Slew Rates with Cyclone® 10 GX Devices 7.4.2. Channel Signal Integrity Measurement 7.4.3. Layout Approach 7.4.4. Design Layout Guidelines 7.4.5. Package Deskew
7.4.1. Terminations and Slew Rates with Cyclone® 10 GX Devices x
7.4.1.1. Dynamic On-Chip Termination (OCT) in Cyclone® 10 Devices 7.4.1.2. Choosing Terminations on Cyclone® 10 Devices 7.4.1.3. On-Chip Termination Recommendations for Cyclone® 10 Devices
7.4.2. Channel Signal Integrity Measurement x
7.4.2.1. Importance of Accurate Channel Signal Integrity Information 7.4.2.2. Understanding Channel Signal Integrity Measurement 7.4.2.3. How to Enter Calculated Channel Signal Integrity Values 7.4.2.4. Guidelines for Calculating DDR3 Channel Signal Integrity
7.4.4. Design Layout Guidelines x
7.4.4.1. General Layout Guidelines 7.4.4.2. Layout Guidelines 7.4.4.3. Length Matching Rules 7.4.4.4. Spacing Guidelines 7.4.4.5. Layout Guidelines for DDR3 SDRAM Wide Interface (>72 bits) 7.4.4.6. Fly-By Network Design for Clock, Command, and Address Signals
7.4.5. Package Deskew x
7.4.5.1. DQ/DQS/DM Deskew 7.4.5.2. Address and Command Deskew 7.4.5.3. Package Deskew Recommendations for Cyclone® 10 Devices 7.4.5.4. Deskew Example 7.4.5.5. Package Migration
8. Cyclone® 10 GX EMIF IP Timing Closure x
8.1. Timing Closure 8.2. Timing Report DDR 8.3. Optimizing Timing 8.4. Early I/O Timing Estimation
8.1. Timing Closure x
8.1.1. Timing Analysis
8.1.1. Timing Analysis x
8.1.1.1. PHY or Core 8.1.1.2. I/O Timing
8.1.1.2. I/O Timing x
8.1.1.2.1. Read Capture 8.1.1.2.2. Write 8.1.1.2.3. Address and Command 8.1.1.2.4. DQS Gating / Postamble 8.1.1.2.5. Write Leveling
8.4. Early I/O Timing Estimation x
8.4.1. Performing Early I/O Timing Analysis
9. Optimizing Controller Performance x
9.1. Interface Standard 9.2. Bank Management Efficiency 9.3. Data Transfer 9.4. Improving Controller Efficiency
9.4. Improving Controller Efficiency x
9.4.1. Auto-Precharge Commands 9.4.2. Latency 9.4.3. Calibration 9.4.4. Bank Interleaving 9.4.5. Additive Latency and Bank Interleaving 9.4.6. User-Controlled Refresh 9.4.7. Frequency of Operation 9.4.8. Series of Reads or Writes 9.4.9. Data Reordering 9.4.10. Starvation Control 9.4.11. Command Reordering 9.4.12. Bandwidth 9.4.13. Enable Command Priority Control
9.4.2. Latency x
9.4.2.1. Additive Latency
9.4.6. User-Controlled Refresh x
9.4.6.1. Back-to-Back User-Controlled Refresh Usage
10. Cyclone® 10 GX EMIF IP Debugging x
10.1. Interface Configuration Performance Issues 10.2. Functional Issue Evaluation 10.3. Timing Issue Characteristics 10.4. Verifying Memory IP Using the Signal Tap II Logic Analyzer 10.5. Hardware Debugging Guidelines 10.6. Categorizing Hardware Issues 10.7. Debugging Cyclone® 10 GX EMIF IP 10.8. Using the Traffic Generator with the Generated Design Example
10.1. Interface Configuration Performance Issues x
10.1.1. Interface Configuration Bottleneck and Efficiency Issues
10.2. Functional Issue Evaluation x
10.2.1. Intel® IP Memory Model 10.2.2. Vendor Memory Model 10.2.3. Transcript Window Messages 10.2.4. Modifying the Example Driver to Replicate the Failure
10.3. Timing Issue Characteristics x
10.3.1. Evaluating FPGA Timing Issues 10.3.2. Evaluating External Memory Interface Timing Issues
10.4. Verifying Memory IP Using the Signal Tap II Logic Analyzer x
10.4.1. Signals to Monitor with the Signal Tap II Logic Analyzer
10.5. Hardware Debugging Guidelines x
10.5.1. Create a Simplified Design that Demonstrates the Same Issue 10.5.2. Measure Power Distribution Network 10.5.3. Measure Signal Integrity and Setup and Hold Margin 10.5.4. Vary Voltage 10.5.5. Operate at a Lower Speed 10.5.6. Determine Whether the Issue Exists in Previous Versions of Software 10.5.7. Determine Whether the Issue Exists in the Current Version of Software 10.5.8. Try A Different PCB 10.5.9. Try Other Configurations 10.5.10. Debugging Checklist
10.6. Categorizing Hardware Issues x
10.6.1. Signal Integrity Issues 10.6.2. Hardware and Calibration Issues
10.6.1. Signal Integrity Issues x
10.6.1.1. Characteristics of Signal Integrity Issues 10.6.1.2. Evaluating SignaI Integrity Issues
10.6.1.2. Evaluating SignaI Integrity Issues x
10.6.1.2.1. Skew 10.6.1.2.2. Crosstalk 10.6.1.2.3. Power System 10.6.1.2.4. Clock Signals 10.6.1.2.5. Read Data Valid Window and Eye Diagram 10.6.1.2.6. Write Data Valid Window and Eye Diagram 10.6.1.2.7. OCT and ODT Usage
10.6.2. Hardware and Calibration Issues x
10.6.2.1. Postamble Timing Issues and Margin 10.6.2.2. Intermittent Issue Evaluation
10.7. Debugging Cyclone® 10 GX EMIF IP x
10.7.1. Debugging With the External Memory Interface Debug Toolkit 10.7.2. On-Chip Debug Port for Cyclone® 10 GX EMIF IP 10.7.3. Efficiency Monitor and Protocol Checker
10.7.1. Debugging With the External Memory Interface Debug Toolkit x
10.7.1.1. User Interface 10.7.1.2. Communication 10.7.1.3. Setup and Use 10.7.1.4. Configuring Your EMIF IP for Use with the Debug Toolkit 10.7.1.5. Reports 10.7.1.6. Driver Margining for Cyclone® 10 GX EMIF IP 10.7.1.7. Example Tcl Script for Running the EMIF Debug Toolkit
10.7.1.4. Configuring Your EMIF IP for Use with the Debug Toolkit x
10.7.1.4.1. Daisy-Chaining Additional EMIF IP Cores for Debugging 10.7.1.4.2. General Workflow 10.7.1.4.3. Linking the Project to a Device 10.7.1.4.4. Establishing Communication to Connections 10.7.1.4.5. Selecting an Active Interface
10.7.2. On-Chip Debug Port for Cyclone® 10 GX EMIF IP x
10.7.2.1. EMIF On-Chip Debug Port 10.7.2.2. Access Protocol
10.7.3. Efficiency Monitor and Protocol Checker x
10.7.3.1. Including the Efficiency Monitor and Protocol Checker in Your Generated IP 10.7.3.2. Running the Efficiency Monitor with the External Memory Debug Toolkit 10.7.3.3. Communicating Directly to the Efficiency Monitor and Protocol Checker
1. Release Information
2. External Memory Interfaces Cyclone® 10 GX FPGA IP Introduction
3. Cyclone® 10 GX EMIF IP Product Architecture
4. Cyclone® 10 GX EMIF IP End-User Signals
5. Cyclone® 10 GX EMIF – Simulating Memory IP
6. Cyclone® 10 GX EMIF IP for DDR3
7. Cyclone® 10 GX EMIF IP for LPDDR3
8. Cyclone® 10 GX EMIF IP Timing Closure
9. Optimizing Controller Performance
10. Cyclone® 10 GX EMIF IP Debugging
11. External Memory Interfaces Cyclone® 10 GX FPGA IP User Guide Archives
12. Document Revision History for Cyclone® 10 GX External Memory Interfaces FPGA IP User Guide

6.4.4.2. Layout Guidelines

The following table lists layout guidelines.

Unless otherwise specified, the guidelines in the following table apply to the following topologies:

  • DIMM—UDIMM topology
  • DIMM—RDIMM topology
  • DIMM—LRDIMM topology
  • Not all versions of the Quartus® Prime software support LRDIMM.
  • Discrete components laid out in UDIMM topology
  • Discrete components laid out in RDIMM topology

These guidelines are recommendations, and should not be considered as hard requirements. You should perform signal integrity simulation on all the traces to verify the signal integrity of the interface.

For supported frequencies and topologies, refer to the External Memory Interface Spec Estimator https://www.intel.com/content/www/us/en/programmable/support/support-resources/external-memory.html.

For frequencies greater than 800 MHz, when you are calculating the delay associated with a trace, you must take the FPGA package delays into consideration.

Table 117.  Layout Guidelines   (1)

Parameter

Guidelines

Decoupling Parameter

  • Make VTT voltage decoupling close to the components and pull-up resistors.
  • Connect decoupling caps between VTT and VDD using a 0.1F cap for every other VTT pin.
  • Use a 0.1 uF cap and 0.01 uF cap for every VDDQ pin.

Maximum Trace Length

  • Even though there are no hard requirements for minimum trace length, you need to simulate the trace to ensure the signal integrity. Shorter routes result in better timing.
  • For DIMM topology only:
    • Maximum trace length for all signals from FPGA to the first DIMM slot is 4.5 inches.
    • Maximum trace length for all signals from DIMM slot to DIMM slot is 0.425 inches.
  • For discrete components only:
    • Maximum trace length for address, command, control, and clock from FPGA to the first component must not be more than 7 inches.
    • Maximum trace length for DQ, DQS, DQS#, and DM from FPGA to the first component is 5 inches.

General Routing

  • Route over appropriate VCC and GND planes.
  • Keep signal routing layers close to GND and power planes.
Spacing Guidelines
  • Avoid routing two signal layers next to each other. Always make sure that the signals related to memory interface are routed between appropriate GND or power layers.
  • For DQ/DQS/DM traces: Maintain at least 3H spacing between the edges (air-gap) for these traces. (Where H is the vertical distance to the closest return path for that particular trace.)
  • For Address/Command/Control traces: Maintain at least 3H spacing between the edges (air-gap) these traces. (Where H is the vertical distance to the closest return path for that particular trace.)
  • For Clock traces: Maintain at least 5H spacing between two clock pair or a clock pair and any other memory interface trace. (Where H is the vertical distance to the closest return path for that particular trace.)

Clock Routing

  • Route clocks on inner layers with outer-layer run lengths held to under 500 mils (12.7 mm).
  • Route clock signals in a daisy chain topology from the first SDRAM to the last SDRAM. The maximum length of the first SDRAM to the last SDRAM must not exceed 0.69 tCK for DDR3 . For different DIMM configurations, check the appropriate JEDEC* specification.
  • These signals should maintain the following spacings:
  • Clocks should maintain a length-matching between clock pairs of ±5 ps.
  • Clocks should maintain a length-matching between positive (p) and negative (n) signals of ±2 ps, routed in parallel.
  • Space between different pairs should be at least two times the trace width of the differential pair to minimize loss and maximize interconnect density.
  • To avoid mismatched transmission line to via, Intel® recommends that you use Ground Signal Signal Ground (GSSG) topology for your clock pattern—GND|CLKP|CKLN|GND.
  • Route all addresses and commands to match the clock signals to within ±20 ps to each discrete memory component. Refer to the following figure.

Address and Command Routing

  • Route address and command signals in a daisy chain topology from the first SDRAM to the last SDRAM. The maximum length of the first SDRAM to the last SDRAM must not be more than 0.69 tCK for DDR3 . For different DIMM configurations, check the appropriate JEDEC* specifications.
  • UDIMMs are more susceptible to cross-talk and are generally noisier than buffered DIMMs. Therefore, route address and command signals of UDIMMs on a different layer than data signals (DQ) and data mask signals (DM) and with greater spacing.
  • Do not route differential clock (CK) and clock enable (CKE) signals close to address signals.
  • Route all addresses and commands to match the clock signals to within ±20 ps to each discrete memory component. Refer to the following figure.

DQ, DM, and DQS Routing Rules

  • All the trace length matching requirements are from the FPGA package ball to the SDRAM package ball, which means you must consider trace mismatching on different DIMM raw cards.
  • Match in length all DQ, DQS, and DM signals within a given byte-lane group with a maximum deviation of ±10 ps.
  • Ensure to route all DQ, DQS, and DM signals within a given byte-lane group on the same layer to avoid layer to layer transmission velocity differences, which otherwise increase the skew within the group.
  • Do not count on FPGAs to deskew for more than 20 ps of DQ group skew. The skew algorithm only removes the following possible uncertainties:
    • Minimum and maximum die IOE skew or delay mismatch
    • Minimum and maximum device package skew or mismatch
    • Board delay mismatch of 20 ps
    • Memory component DQ skew mismatch
    • Increasing any of these four parameters runs the risk of the deskew algorithm limiting, failing to correct for the total observed system skew. If the algorithm cannot compensate without limiting the correction, timing analysis shows reduced margins.
  • For memory interfaces with leveling, the timing between the DQS and clock signals on each device calibrates dynamically to meet tDQSS. To make sure the skew is not too large for the leveling circuit’s capability, follow these rules:
    • Propagation delay of clock signal must not be shorter than propagation delay of DQS signal at every device: (CKi) – DQSi > 0; 0 < i < number of components – 1 . For DIMMs, ensure that the CK trace is longer than the longest DQS trace at the DIMM connector.
    • Total skew of CLK and DQS signal between groups is less than one clock cycle: (CKi+ DQSi) max – (CKi+ DQSi) min < 1 × tCK(If you are using a DIMM topology, your delay and skew must take into consideration values for the actual DIMM.)

Spacing Guidelines
  • Avoid routing two signal layers next to each other. Always ensure that the signals related to the memory interface are routed between appropriate GND or power layers.
  • For DQ/DQS/DM traces: Maintain at least 3H spacing between the edges (air-gap) of these traces, where H is the vertical distance to the closest return path for that particular trace.
  • For Address/Command/Control traces: Maintain at least 3H spacing between the edges (air-gap) of these traces, where H is the vertical distance to the closest return path for that particular trace.
  • For Clock traces: Maintain at least 5H spacing between two clock pairs or a clock pair and any other memory interface trace, where H is the vertical distance to the closest return path for that particular trace.

Quartus® Prime Software Settings for Board Layout

  • To perform timing analyses on board and I/O buffers, use a third-party simulation tool to simulate all timing information such as skew, ISI, crosstalk, and type the simulation result into the Board Settings tab in the parameter editor.
  • Do not use advanced I/O timing model (AIOT) or board trace model unless you do not have access to any third party tool. AIOT provides reasonable accuracy but tools like HyperLynx provide better results.

Notes to Table:

  1. For point-to-point and DIMM interface designs, refer to the Micron website, www.micron.com.
Related Information
Level Two Title