PHY Lite for Parallel Interfaces Intel® FPGA IP User Guide

Download PDF
ID 683716
Date 9/01/2021
Public

A newer version of this document is available.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

Document Table of Contents
Document Table of Contents x
1. About the PHY Lite for Parallel Interfaces IP 2. PHY Lite for Parallel Interfaces Intel Agilex FPGA IP 3. PHY Lite for Parallel Interfaces Intel Stratix 10 FPGA IP 4. PHY Lite for Parallel Interfaces Intel Arria 10 and Intel Cyclone 10 GX FPGA IPs 5. PHY Lite for Parallel Interfaces Intel® FPGA IP User Guide Document Archives 6. Document Revision History for the PHY Lite for Parallel Interfaces IP User Guide
1. About the PHY Lite for Parallel Interfaces IP x
1.1. Device Family Support 1.2. Features
2. PHY Lite for Parallel Interfaces Intel Agilex FPGA IP x
2.1. Release Information 2.2. Functional Description 2.3. Getting Started 2.4. I/O Standards 2.5. Design Guidelines 2.6. Design Example
2.2. Functional Description x
2.2.1. Intel® Agilex™ I/O Sub-bank Interconnects 2.2.2. Intel® Agilex™ Input DQS/Strobe Tree 2.2.3. PHY Lite for Parallel Interfaces Intel® Agilex™ FPGA IP Top Level Interfaces 2.2.4. Dynamic Reconfiguration 2.2.5. I/O Timing
2.2.3. PHY Lite for Parallel Interfaces Intel® Agilex™ FPGA IP Top Level Interfaces x
2.2.3.1. Clocks 2.2.3.2. Output Path 2.2.3.3. Input Path
2.2.3.1. Clocks x
2.2.3.1.1. Clock Frequency Relationships
2.2.3.2. Output Path x
2.2.3.2.1. Output Path Data Alignment
2.2.3.3. Input Path x
2.2.3.3.1. Input Path Data Alignment
2.2.4. Dynamic Reconfiguration x
2.2.4.1. Connectivity 2.2.4.2. Reconfiguration Features and Register Addressing 2.2.4.3. Dynamic Reconfiguration Guidelines
2.2.4.2. Reconfiguration Features and Register Addressing x
2.2.4.2.1. Control Registers Addresses 2.2.4.2.2. Control Registers
2.2.4.3. Dynamic Reconfiguration Guidelines x
2.2.4.3.1. Strobe Enable Window Calibration 2.2.4.3.2. Output and Strobe Enable Minimum and Maximum Phase Settings 2.2.4.3.3. Input DQ/DQS Delay Chains Maximum Values
2.3. Getting Started x
2.3.1. Parameter Settings 2.3.2. Signals
2.3.1. Parameter Settings x
2.3.1.1. Read Latency 2.3.1.2. Write Latency
2.3.2. Signals x
2.3.2.1. Clock and Reset Interface Signals 2.3.2.2. Output Path Signals 2.3.2.3. Input Path Signals
2.4. I/O Standards x
2.4.1. Input Buffer Reference Voltage (VREF) 2.4.2. On-Chip Termination (OCT)
2.4.2. On-Chip Termination (OCT) x
2.4.2.1. RZQ_GROUP Assignment
2.5. Design Guidelines x
2.5.1. Guidelines: Group Pin Placement 2.5.2. Reference Clock 2.5.3. Reset
2.6. Design Example x
2.6.1. Generate the Design Example
2.6.1. Generate the Design Example x
2.6.1.1. Design Example without Dynamic Reconfiguration 2.6.1.2. Dynamic Reconfiguration Design Examples
2.6.1.1. Design Example without Dynamic Reconfiguration x
2.6.1.1.1. Generate the Synthesis Design Example 2.6.1.1.2. Generate the Simulation Design Example
2.6.1.2. Dynamic Reconfiguration Design Examples x
2.6.1.2.1. Generate the Synthesis Design Example 2.6.1.2.2. Generate the Simulation Design Example
3. PHY Lite for Parallel Interfaces Intel Stratix 10 FPGA IP x
3.1. Release Information 3.2. Functional Description 3.3. Getting Started 3.4. I/O Standards 3.5. Design Guidelines 3.6. Design Example
3.2. Functional Description x
3.2.1. Top Level Interfaces 3.2.2. Clocks 3.2.3. Output Path 3.2.4. Input Path 3.2.5. Dynamic Reconfiguration
3.2.2. Clocks x
3.2.2.1. Clock Frequency Relationships
3.2.3. Output Path x
3.2.3.1. Output Path Data Alignment
3.2.4. Input Path x
3.2.4.1. Input Path Data Alignment
3.2.5. Dynamic Reconfiguration x
3.2.5.1. RTL Connectivity 3.2.5.2. Address Lookup 3.2.5.3. Reconfiguration Features and Register Addressing 3.2.5.4. Calibration Guidelines
3.2.5.1. RTL Connectivity x
3.2.5.1.1. Daisy Chain
3.2.5.2. Address Lookup x
3.2.5.2.1. Strobes 3.2.5.2.2. Parameter Table Example
3.2.5.3. Reconfiguration Features and Register Addressing x
3.2.5.3.1. PHY Lite for Parallel Interfaces Intel Stratix 10 FPGA IP Control Registers Addresses 3.2.5.3.2. Control Registers Description
3.2.5.4. Calibration Guidelines x
3.2.5.4.1. Strobe Enable Window Calibration 3.2.5.4.2. Output and Strobe Enable Minimum and Maximum Phase Settings
3.3. Getting Started x
3.3.1. Parameter Settings 3.3.2. Signals
3.3.1. Parameter Settings x
3.3.1.1. Read Latency 3.3.1.2. Write Latency
3.3.2. Signals x
3.3.2.1. Clock and Reset Interface Signals 3.3.2.2. Output Path Signals 3.3.2.3. Input Path Signals 3.3.2.4. Avalon Configuration Bus Interface Signals 3.3.2.5. Termination Signals
3.4. I/O Standards x
3.4.1. Input Buffer Reference Voltage (VREF) 3.4.2. On-Chip Termination (OCT)
3.4.1. Input Buffer Reference Voltage (VREF) x
3.4.1.1. Calibrated VREF Settings
3.4.2. On-Chip Termination (OCT) x
3.4.2.1. RZQ_GROUP Assignment 3.4.2.2. Manual Insertion of OCT Block
3.5. Design Guidelines x
3.5.1. Guidelines: Group Pin Placement 3.5.2. Reference Clock 3.5.3. Reset 3.5.4. Constraining Multiple PHY Lite for Parallel Interfaces to One I/O Bank 3.5.5. Dynamic Reconfiguration 3.5.6. Timing
3.5.6. Timing x
3.5.6.1. Timing Components 3.5.6.2. Timing Constraints and Files 3.5.6.3. Timing Analysis 3.5.6.4. Timing Closure Guidelines
3.5.6.2. Timing Constraints and Files x
3.5.6.2.1. <variation_name>.sdc 3.5.6.2.2. <variation_name>_parameter.tcl 3.5.6.2.3. <variation_name>_ip_parameters.tcl 3.5.6.2.4. <variation_name>_pin_map.tcl 3.5.6.2.5. <variation_name>_report_timing.tcl 3.5.6.2.6. <variation_name>_report_timing_core.tcl
3.5.6.4. Timing Closure Guidelines x
3.5.6.4.1. Timing Closure: Dynamic Reconfiguration 3.5.6.4.2. Timing Closure: Input Strobe Setup and Hold Delay Constraints 3.5.6.4.3. Timing Closure: Output Strobe Setup and Hold Delay Constraints 3.5.6.4.4. Timing Closure: Non Edge-Aligned Input Data 3.5.6.4.5. I/O Timing Violation 3.5.6.4.6. Internal FPGA Path Timing Violation
3.6. Design Example x
3.6.1. Generate the Design Example
3.6.1. Generate the Design Example x
3.6.1.1. Design Example without Dynamic Reconfiguration 3.6.1.2. Dynamic Reconfiguration Design Examples
3.6.1.1. Design Example without Dynamic Reconfiguration x
3.6.1.1.1. Generate the Hardware Design Example 3.6.1.1.2. Generate the Simulation Design Example
3.6.1.2. Dynamic Reconfiguration Design Examples x
3.6.1.2.1. Dynamic Reconfiguration Using Finite State Machine
4. PHY Lite for Parallel Interfaces Intel Arria 10 and Intel Cyclone 10 GX FPGA IPs x
4.1. Release Information 4.2. Functional Description 4.3. Getting Started 4.4. I/O Standards 4.5. Design Guidelines 4.6. Design Example 4.7. Application Specific Design Example
4.2. Functional Description x
4.2.1. Top Level Interfaces 4.2.2. Clocks 4.2.3. Output Path 4.2.4. Input Path 4.2.5. Dynamic Reconfiguration
4.2.2. Clocks x
4.2.2.1. Clock Frequency Relationships
4.2.3. Output Path x
4.2.3.1. Output Path Data Alignment
4.2.4. Input Path x
4.2.4.1. Input Path Data Alignment
4.2.5. Dynamic Reconfiguration x
4.2.5.1. RTL Connectivity 4.2.5.2. Address Lookup 4.2.5.3. Reconfiguration Features and Register Addressing 4.2.5.4. Calibration Guidelines
4.2.5.1. RTL Connectivity x
4.2.5.1.1. Daisy Chain
4.2.5.2. Address Lookup x
4.2.5.2.1. Strobes 4.2.5.2.2. Parameter Table Example
4.2.5.3. Reconfiguration Features and Register Addressing x
4.2.5.3.1. PHY Lite for Parallel Interfaces Intel Arria 10 FPGA IP and PHY Lite for Parallel Interfaces Intel Cyclone 10 GX IPs Address Registers 4.2.5.3.2. Control Registers Description
4.2.5.4. Calibration Guidelines x
4.2.5.4.1. Strobe Enable Window Calibration 4.2.5.4.2. Output and Strobe Enable Minimum and Maximum Phase Settings
4.3. Getting Started x
4.3.1. Parameter Settings 4.3.2. Signals
4.3.1. Parameter Settings x
4.3.1.1. Read Latency 4.3.1.2. Write Latency
4.3.2. Signals x
4.3.2.1. Clock and Reset Interface Signals 4.3.2.2. Output Path Signals 4.3.2.3. Input Path Signals 4.3.2.4. Avalon Configuration Bus Interface Signals 4.3.2.5. Termination Signals
4.4. I/O Standards x
4.4.1. Input Buffer Reference Voltage (VREF) 4.4.2. On-Chip Termination (OCT)
4.4.1. Input Buffer Reference Voltage (VREF) x
4.4.1.1. Calibrated VREF Settings
4.4.2. On-Chip Termination (OCT) x
4.4.2.1. RZQ_GROUP Assignment 4.4.2.2. Manual Insertion of OCT Block
4.5. Design Guidelines x
4.5.1. Guidelines: Group Pin Placement 4.5.2. Reference Clock 4.5.3. Reset 4.5.4. Constraining Multiple PHY Lite for Parallel Interfaces to One I/O Bank 4.5.5. Dynamic Reconfiguration 4.5.6. Timing
4.5.6. Timing x
4.5.6.1. Timing Components 4.5.6.2. Timing Constraints and Files 4.5.6.3. Timing Analysis 4.5.6.4. Timing Closure Guidelines
4.5.6.2. Timing Constraints and Files x
4.5.6.2.1. <variation_name>.sdc 4.5.6.2.2. <variation_name>_parameter.tcl 4.5.6.2.3. <variation_name>_ip_parameters.tcl 4.5.6.2.4. <variation_name>_pin_map.tcl 4.5.6.2.5. <variation_name>_report_timing.tcl 4.5.6.2.6. <variation_name>_report_timing_core.tcl
4.5.6.4. Timing Closure Guidelines x
4.5.6.4.1. Timing Closure: Dynamic Reconfiguration 4.5.6.4.2. Timing Closure: Input Strobe Setup and Hold Delay Constraints 4.5.6.4.3. Timing Closure: Output Strobe Setup and Hold Delay Constraints 4.5.6.4.4. Timing Closure: Non Edge-Aligned Input Data 4.5.6.4.5. I/O Timing Violation 4.5.6.4.6. Internal FPGA Path Timing Violation
4.6. Design Example x
4.6.1. Generate the Design Example
4.6.1. Generate the Design Example x
4.6.1.1. Design Example without Dynamic Reconfiguration 4.6.1.2. Dynamic Reconfiguration Design Examples
4.6.1.1. Design Example without Dynamic Reconfiguration x
4.6.1.1.1. Generate the Hardware Design Example 4.6.1.1.2. Generate the Simulation Design Example
4.6.1.2. Dynamic Reconfiguration Design Examples x
4.6.1.2.1. Dynamic Reconfiguration with Debug Kit 4.6.1.2.2. Dynamic Reconfiguration Using Finite State Machine
4.7. Application Specific Design Example x
4.7.1. Implementation using the PHY Lite for Parallel Interfaces IP
1. About the PHY Lite for Parallel Interfaces IP
2. PHY Lite for Parallel Interfaces Intel Agilex FPGA IP
3. PHY Lite for Parallel Interfaces Intel Stratix 10 FPGA IP
4. PHY Lite for Parallel Interfaces Intel Arria 10 and Intel Cyclone 10 GX FPGA IPs
5. PHY Lite for Parallel Interfaces Intel® FPGA IP User Guide Document Archives
6. Document Revision History for the PHY Lite for Parallel Interfaces IP User Guide

2.2.4.3.1. Strobe Enable Window Calibration

The main block, pstamble_reg, has a gating circuitry to generate dqs_clean from strobe_in/strobe_io and dqs_enable_out signals. The pstamble_reg block has an finite state machine (FSM) to control the gate signal. If dqs_enable_out is high and strobe_in/strobe_io is at any positive edge, the gate signal is asserted and dqs_clean is generated. The dqs_enable_out is high if you set the external signal, rdata_en to high.

An internal counter counts the number of strobe_in/strobe_io toggles until the internal counter reaches the maximum number of toggles. The maximum number of toggles depends on the internal counter. In this example, the internal counter counts eight toggles. Ideally, in normal PHY state, after the maximum number of toggles is reached, the gate signal is deasserted.

Figure 15. Input Path of PHY Lite for Parallel Interfaces Intel® Agilex™ FPGA IP Core

Perform the strobe enable window calibration to capture the correct data­in/data_io on strobe_in/strobe_io signal. You can perform the calibration either by sweeping the dqs_enable_out through the interpolator or sweeping the strobe_in/strobe_io or data_in/data_io.

However, during strobe enable window calibration, while finding the end of the window, an undesired state can happen where the gate signal remains asserted incorrectly and the wrong data­in/data_io (noise) is captured on strobe_in/strobe_io signal. In this case, you must bring PHY to normal state by adding dummy pulses on the strobe_in/strobe_io signal with rdata_en deasserted.

In the following figure, in normal PHY state, the dqs_enable_out is high after the rdata_en signal is high. Once dqs_enable_out is high, if a positive edge of strobe_in/strobe_io is observed, the gate signal goes high (marked by blue marker). An internal counter starts to count at first positive edge of strobe_in/strobe_io until the maximum number of toggles (in this case, the maximum number of toggles is eight). The gate signal is deasserted on the last negative edge of the strobe_in/strobe_io (marked by the green marker). Ideally, dqs_clean is the same as strobe_in/strobe_io because all eight toggles are captured by the internal counter.

In the undesirable state, the strobe_io/strobe_in is calibrated to start toggling earlier. While detecting the correct window margins during calibration, an undesired state occurs because the internal counter does not finish counting to eight resulting in the gate signal remains asserted and produces incomplete dqs_clean (marked by red marker).

Figure 16. dqs_clean Timing Diagram

To return the PHY to the normal state, you must force the gate signal to deassert by adding extra dummy pulses (red strobe_in/strobe_io pulses) while rdata_en is deasserted as shown in the following figure. By adding these dummy pulses, a complete dqs_clean is produced (marked by green marker). After the green marker, although there are few strobe_in/strobe_io toggles, no new dqs_clean is produced because the gate signal remains low.

Figure 17. Adding Extra Dummy Pulses to Return PHY to Normal State
Level Two Title