E-Tile Transceiver PHY User Guide

Download PDF
ID 683723
Date 7/08/2024
Public
Document Table of Contents
Document Table of Contents x
1. E-Tile Transceiver PHY Overview 2. Implementing the Transceiver PHY Layer 3. E-Tile Transceiver PHY Architecture 4. Clock Network 5. PMA Calibration 6. Resetting Transceiver Channels 7. Dynamic Reconfiguration 8. Dynamic Reconfiguration Examples 9. Register Map 10. Debugging E-Tile Transceiver Links A. E-Tile Channel Placement Tool B. PMA Direct PAM4 30 Gbps to 57.8 Gbps Implementation C. Signal Detect Algorithm D. Detailed Steps for Reconfiguring from Mission Mode to Channel Protection Mode E. Detailed Steps for Reconfiguring from Channel Protection Mode to Mission Mode F. Hold Timing Violation
1. E-Tile Transceiver PHY Overview x
1.1. Supported Features 1.2. E-Tile Layout in Stratix® 10 Device Variants 1.3. E-Tile Layout in Agilex™ 7 F-Series Device Variants 1.4. Transceiver Counts 1.5. E-Tile Building Blocks 1.6. E-Tile Transceiver PHY Overview Revision History
1.2. E-Tile Layout in Stratix® 10 Device Variants x
1.2.1. Stratix® 10 TX H-Tile and E-Tile Configurations 1.2.2. Stratix® 10 MX H-Tile and E-Tile Configurations 1.2.3. Stratix® 10 DX P-Tile and E-Tile Configurations
1.5. E-Tile Building Blocks x
1.5.1. GXE Transceiver Channel 1.5.2. GXE Channel Usage 1.5.3. Reference Clocks 1.5.4. Ethernet Hard IP (EHIP) 1.5.5. Supported Applications/Modes 1.5.6. Feature Comparison Between Transceiver Tiles
2. Implementing the Transceiver PHY Layer x
2.1. Transceiver Design Flow in the Native PHY IP Core 2.2. Configuring the Native PHY IP Core 2.3. Implementing the Transceiver PHY Layer Revision History
2.1. Transceiver Design Flow in the Native PHY IP Core x
2.1.1. E-Tile Native PHY IP Core
2.2. Configuring the Native PHY IP Core x
2.2.1. General and Datapath Parameters 2.2.2. PMA Parameters 2.2.3. Core Interface Options 2.2.4. PMA Interface 2.2.5. PMA Adaptation Parameters 2.2.6. Reed Solomon Forward Error Correction (RS-FEC) Parameters 2.2.7. Reset Parameters 2.2.8. Dynamic Reconfiguration Parameters 2.2.9. Deskew Logic 2.2.10. Port Information 2.2.11. PLL Mode 2.2.12. Simplex Support
2.2.2. PMA Parameters x
2.2.2.1. TX PMA Options 2.2.2.2. TX PMA Pre-equalization 2.2.2.3. RX PMA Options 2.2.2.4. RX PMA Optional Ports
2.2.3. Core Interface Options x
2.2.3.1. Core Interface Parameters 2.2.3.2. TX Clock Options 2.2.3.3. RX Clock Options 2.2.3.4. Latency Measurement Options
2.2.4. PMA Interface x
2.2.4.1. PMA Interface Options 2.2.4.2. Gearbox 64/66
2.2.6. Reed Solomon Forward Error Correction (RS-FEC) Parameters x
2.2.6.1. Fibre-Channel and CPRI Modes 2.2.6.2. 128 GFC Mode 2.2.6.3. 25 GbE FEC Direct Mode 2.2.6.4. Interlaken Mode
3. E-Tile Transceiver PHY Architecture x
3.1. Physical Medium Attachment (PMA) Architecture 3.2. Physical Coding Sublayer (PCS) Architecture 3.3. Reed Solomon Forward Error Correction (RS-FEC) Architecture 3.4. E-Tile Transceiver PHY Architecture Revision History
3.1. Physical Medium Attachment (PMA) Architecture x
3.1.1. Transmitter PMA 3.1.2. Receiver PMA 3.1.3. PMA Tuning 3.1.4. Duplex Adaptation Flow 3.1.5. RX Simplex Adaptation Flow 3.1.6. Dynamic Reconfiguration Adaptation Flow 3.1.7. Loopback modes 3.1.8. PMA Interface 3.1.9. TX PMA Bonding 3.1.10. Unused Transceiver Channels 3.1.11. Low Power Mode (LPM)
3.1.1. Transmitter PMA x
3.1.1.1. High Speed Differential Transmitter 3.1.1.2. Gray Encoder/Precoder 3.1.1.3. Serializer 3.1.1.4. Data Pattern Generation
3.1.1.1. High Speed Differential Transmitter x
3.1.1.1.1. High Speed Transmitter Line Buffer 3.1.1.1.2. TX Equalizer
3.1.2. Receiver PMA x
3.1.2.1. Receiver Buffer 3.1.2.2. Clock Data Recovery (CDR) Block 3.1.2.3. Input Sampler 3.1.2.4. Deserializer 3.1.2.5. Data Pattern Verifier
3.1.2.1. Receiver Buffer x
3.1.2.1.1. Programmable Termination Modes 3.1.2.1.2. RX Adaptation Modes
3.1.3. PMA Tuning x
3.1.3.1. Purpose of PMA Tuning 3.1.3.2. PMA Bring Up Flow 3.1.3.3. PMA Tuning Guidelines 3.1.3.4. General PMA Tuning Guidelines
3.1.7. Loopback modes x
3.1.7.1. Internal Serial Loopback Path 3.1.7.2. Reverse Parallel Loopback Path
3.1.9. TX PMA Bonding x
3.1.9.1. Transceiver Interface Deskew Logic 3.1.9.2. Dedicated Balanced PLL Reference Clock Tree
3.1.10. Unused Transceiver Channels x
3.1.10.1. Unused Transceiver Channels in a Used Tile 3.1.10.2. Unused Transceiver Channels in Completely Unused Tiles 3.1.10.3. Unused Transceiver Channels in High-Speed PAM4 Mode 3.1.10.4. Reconfiguring from Mission Mode to Channel Protection Mode 3.1.10.5. Reconfiguring from Channel Protection Mode to Mission Mode
3.3. Reed Solomon Forward Error Correction (RS-FEC) Architecture x
3.3.1. RS-FEC Modes
4. Clock Network x
4.1. Reference Clock Pins 4.2. Core Clock Network Use Case 4.3. Clock Network Revision History
4.1. Reference Clock Pins x
4.1.1. QSF Assignments for Reference Clock Pins 4.1.2. Dynamic Reconfiguration of Reference Clock
4.2. Core Clock Network Use Case x
4.2.1. Single 25 Gbps PMA Direct Channel (with FEC) Within a Single FEC Block 4.2.2. Single 10 Gbps PMA Direct Channel (without FEC) 4.2.3. Four 25 Gbps PMA Direct Channel (with FEC) within a Single FEC Block 4.2.4. PMA Direct 25 Gbps x 4 (FEC Off) 4.2.5. PMA Direct 10.3125 Gbps x 4 4.2.6. PMA Direct 100GE Gbps (25 Gbps x 4) (FEC On) 4.2.7. PMA Direct 100GE PAM4 (50 Gbps x 2) (Aggregate FEC On) 4.2.8. PMA Direct High Data Rate (FEC Off)
4.2.3. Four 25 Gbps PMA Direct Channel (with FEC) within a Single FEC Block x
4.2.3.1. Master-Slave Configuration: Option 1 4.2.3.2. Master-Slave Configuration: Option 2
5. PMA Calibration x
5.1. PMA Calibration Revision History
6. Resetting Transceiver Channels x
6.1. When Is Reset Required? 6.2. How Do I Reset? 6.3. Reset Block Architecture 6.4. PMA Analog Reset 6.5. High Level Specification 6.6. Master-Slave Clocking Option 2 Reset Details 6.7. Quartus® Prime Instantiated Transceiver Reset Sequencer 6.8. Block Diagrams 6.9. Interfaces 6.10. Resetting Transceiver Channels Revision History
6.2. How Do I Reset? x
6.2.1. Disabling the E-Tile Transceiver 6.2.2. Selecting the Reset Controller's Clock Source
6.5. High Level Specification x
6.5.1. Automatic Reset Mode 6.5.2. Manual Reset Mode 6.5.3. Reset Controller Bypass
6.5.3. Reset Controller Bypass x
6.5.3.1. Reset Controller Bypass Ports 6.5.3.2. Reset Controller Bypass Reset Procedure
6.9. Interfaces x
6.9.1. Reset Parameters in the Native PHY GUI 6.9.2. HDL Ports/Interfaces
7. Dynamic Reconfiguration x
7.1. Dynamically Reconfiguring Channel Blocks 7.2. Dynamic Reconfiguration Maximum Data Rate Switch 7.3. Interacting with the Dynamic Reconfiguration Interface 7.4. Unsupported Features 7.5. Reading from the Dynamic Reconfiguration Interface 7.6. Writing to the Dynamic Reconfiguration Interface 7.7. Multiple Reconfiguration Profiles 7.8. Arbitration 7.9. Recommendations for PMA Dynamic Reconfiguration 7.10. Steps to Perform Dynamic Reconfiguration 7.11. PMA Attribute Details 7.12. Dynamic Reconfiguration Flow for Special Cases 7.13. Ports and Parameters 7.14. Embedded Debug Features 7.15. Timing Closure Recommendations 7.16. Transceiver Register Map 7.17. Loading IP Configuration Settings 7.18. Dynamic Reconfiguration Revision History
7.7. Multiple Reconfiguration Profiles x
7.7.1. Reconfiguration Files 7.7.2. Embedded Reconfiguration Streamer
7.7.2. Embedded Reconfiguration Streamer x
7.7.2.1. Register 0x40140 7.7.2.2. Register 0x40141
7.12. Dynamic Reconfiguration Flow for Special Cases x
7.12.1. Switching Reference Clocks 7.12.2. Reconfiguring PMA Settings
7.12.1. Switching Reference Clocks x
7.12.1.1. Switching Between Any Two refclk[0, 1, 2, 3, 4, 5, 6, 7] Reference Clocks or Changing the Reference Clock Frequency on refclk[0, 2, 3, 4, 5, 6, 7, 8] 7.12.1.2. Switching from/to refclk[0, 1, 2, 3, 4, 5, 6, 7] to/from refclk[8] 7.12.1.3. Changing the refclk[1] Reference Clock Frequency
7.14. Embedded Debug Features x
7.14.1. Native PHY Debug Master Endpoint (NPDME) 7.14.2. Optional Dynamic Reconfiguration Logic
7.14.2. Optional Dynamic Reconfiguration Logic x
7.14.2.1. Capability Registers 7.14.2.2. Control and Status Registers
7.17. Loading IP Configuration Settings x
7.17.1. Loading IP Configuration Settings Process 7.17.2. Alternative Method for Setting PMA Attributes
8. Dynamic Reconfiguration Examples x
8.1. Reconfiguring the Duplex PMA Using the Reset Controller in Automatic Mode 8.2. PRBS Usage Model 8.3. PMA Error Injection 8.4. PMA Receiver Equalization Adaptation Usage Model 8.5. User-Defined Pattern Example 8.6. Configuring the Attenuation Value (VOD) 8.7. Configuring the Post Emphasis Value 8.8. Configuring pretap1 Values 8.9. Inverting TX Polarity for the PMA Driver 8.10. Inverting RX Polarity for the PMA Driver 8.11. Configuring a PMA Parameter Tunable by the Adaptive Engine 8.12. Configuring a PMA Parameter Using Native PHY IP 8.13. Enabling Low Power Mode for Multiple Channels 8.14. Initializing an RX 8.15. Resetting the RX Equalization 8.16. Dynamic Reconfiguration Examples Revision History
8.12. Configuring a PMA Parameter Using Native PHY IP x
8.12.1. PMA Bring Up Flow Using Native PHY IP 8.12.2. Native PHY IP GUI Details 8.12.3. Loading a PMA Configuration
9. Register Map x
9.1. PMA Register Map 9.2. PMA Attribute Codes 9.3. PMA Registers 0x200 to 0x203 Usage 9.4. Supported Data Rate Ratios for PMA Attribute Codes 0x0005 and 0x0006 9.5. RS-FEC Registers 9.6. Register Map Revision History
9.1. PMA Register Map x
9.1.1. PMA Capability Registers 9.1.2. PMA Control and Status Registers 9.1.3. PMA Avalon® Memory-Mapped Interface
9.2. PMA Attribute Codes x
9.2.1. 0x0001: PMA Enable/Disable 9.2.2. 0x0002: PMA PRBS Settings 9.2.3. 0x0003: Data Comparison Set Up and Start/Stop 9.2.4. 0x0005: TX Channel Divide By Ratio 9.2.5. 0x0006: RX Channel Divide By Ratio 9.2.6. 0x0008: Internal Serial Loopback and Reverse Parallel Loopback Control 9.2.7. 0x000A: Receiver Tuning Controls 9.2.8. 0x000E: RX Phase Slip 9.2.9. 0x0011: PMA TX/RX Calibration 9.2.10. 0x0013: TX/RX Polarity and Gray Code Encoding 9.2.11. 0x0014: TX/RX Width Mode 9.2.12. 0x0015: TX Equalization 9.2.13. 0x0017: Error Counter Reset 9.2.14. 0x0018: Status/Debug Register 9.2.15. 0x0019: Status/Debug Register Next Write Field 9.2.16. 0x001A: Status/Debug Register Next Read Field 9.2.17. 0x001B: TX Error Injection Signal 9.2.18. 0x001C: Incoming RX Data Capture 9.2.19. 0x001E: Error Count Status 9.2.20. 0x0020: Electrical Idle Detector 9.2.21. 0x002B: RX Termination and TX Driver Tri-state Behavior 9.2.22. 0x0030: PMA Mux Clock Swap 9.2.23. 0x0126: Read Receiver Tuning Parameters 9.2.24. Reading and Writing PMA Analog Parameters Using Attributes
9.2.24. Reading and Writing PMA Analog Parameters Using Attributes x
9.2.24.1. Reading PMA Analog Parameters 9.2.24.2. Updating PMA Analog Parameters 9.2.24.3. Loading Parameters into the Receiver 9.2.24.4. Fixing Parameter Values 9.2.24.5. Reading NRZ/PAM4 Eye Height 9.2.24.6. Enabling and Disabling Electrical Idle Detector Filtering and Reading Electrical Idle Detector Status 9.2.24.7. Initial Adaptation Effort Levels
9.3. PMA Registers 0x200 to 0x203 Usage x
9.3.1. PMA Analog Reset 9.3.2. Set PRBS Mode and Internal Serial Loopback 9.3.3. Use Cases for Opcode START_ADAPTATION 9.3.4. Read the Physical Channel Number 9.3.5. Check the PMA Adaptation Status 9.3.6. Load a PMA Configuration using Opcode LOAD_PMA_CONFIGURATION
9.5. RS-FEC Registers x
9.5.1. rsfec_top_clk_cfg 9.5.2. rsfec_top_tx_cfg 9.5.3. rsfec_top_rx_cfg 9.5.4. tx_aib_dsk_conf 9.5.5. rsfec_core_cfg 9.5.6. rsfec_lane_cfg 9.5.7. tx_aib_dsk_status 9.5.8. rsfec_debug_cfg 9.5.9. rsfec_lane_tx_stat 9.5.10. rsfec_lane_tx_hold 9.5.11. rsfec_lane_tx_inten 9.5.12. rsfec_lane_rx_stat 9.5.13. rsfec_lane_rx_hold 9.5.14. rsfec_lane_rx_inten 9.5.15. rsfec_lanes_rx_stat 9.5.16. rsfec_lanes_rx_hold 9.5.17. rsfec_lanes_rx_inten 9.5.18. rsfec_ln_mapping_rx 9.5.19. rsfec_ln_skew_rx 9.5.20. rsfec_cw_pos_rx 9.5.21. rsfec_core_ecc_hold 9.5.22. rsfec_err_inj_tx 9.5.23. rsfec_err_val_tx 9.5.24. rsfec_corr_cw_cnt (Low) 9.5.25. rsfec_corr_cw_cnt (High) 9.5.26. rsfec_uncorr_cw_cnt (Low) 9.5.27. rsfec_uncorr_cw_cnt (High) 9.5.28. rsfec_corr_syms_cnt (Low) 9.5.29. rsfec_corr_syms_cnt (High) 9.5.30. rsfec_corr_0s_cnt (Low) 9.5.31. rsfec_corr_0s_cnt (High) 9.5.32. rsfec_corr_1s_cnt (Low) 9.5.33. rsfec_corr_1s_cnt (High)
10. Debugging E-Tile Transceiver Links x
10.1. E-Tile Transceiver Toolkit Overview 10.2. E-Tile Transceiver Debugging Flow Walkthrough 10.3. Modifying the Design to Enable E-Tile Transceiver Debug 10.4. Programming the Design into an Intel FPGA 10.5. Loading the Design in the E-Tile Transceiver Toolkit 10.6. Verifying E-Tile Hardware Connections 10.7. Running Transceiver Tests 10.8. Controlling PMA Analog Settings 10.9. Debugging E-Tile Transceiver Links Revision History
10.3. Modifying the Design to Enable E-Tile Transceiver Debug x
10.3.1. Debug Parameters for the E-Tile Transceiver IP in the Parameter Editor 10.3.2. Debug Parameters for Stratix® 10 E-Tile Transceiver Native PHY IP in the Parameter Editor 10.3.3. Instantiating and Parameterizing E-Tile Transceiver IP
10.7. Running Transceiver Tests x
10.7.1. Running BER Tests 10.7.2. Link Optimization Tests 10.7.3. Running Eye Viewer Tests
A. E-Tile Channel Placement Tool x
A.1. E-Tile Channel Placement Tool Revision History
B. PMA Direct PAM4 30 Gbps to 57.8 Gbps Implementation x
B.1. Building Blocks and Considerations B.2. Starting a New Quartus® Prime Pro Edition Design B.3. Selecting the Configuration Clock Source B.4. Instantiating the Transceiver Native PHY IP B.5. Instantiating the In-system Sources and Probes Intel® FPGA IP B.6. Making the Top Level Connection B.7. Assigning Pins B.8. Bringing up the Board B.9. Debug Tools B.10. PMA Direct PAM4 30 Gbps to 57.8 Gbps Implementation Revision History
B.9. Debug Tools x
B.9.1. Monitoring Transceiver Signals
C. Signal Detect Algorithm x
Valid Signal Initial Adaptation Stage Ongoing Adaptation Stage Internal Serial Loopback C.1. Signal Detect Algorithm Revision History
D. Detailed Steps for Reconfiguring from Mission Mode to Channel Protection Mode x
D.1. Detailed Steps for Reconfiguring from Mission Mode to Channel Protection Mode Revision History
E. Detailed Steps for Reconfiguring from Channel Protection Mode to Mission Mode x
E.1. Detailed Steps for Reconfiguring from Channel Protection Mode to Mission Mode Revision History
F. Hold Timing Violation x
F.1. Hold Timing Violation Revision History
1. E-Tile Transceiver PHY Overview
2. Implementing the Transceiver PHY Layer
3. E-Tile Transceiver PHY Architecture
4. Clock Network
5. PMA Calibration
6. Resetting Transceiver Channels
7. Dynamic Reconfiguration
8. Dynamic Reconfiguration Examples
9. Register Map
10. Debugging E-Tile Transceiver Links
A. E-Tile Channel Placement Tool
B. PMA Direct PAM4 30 Gbps to 57.8 Gbps Implementation
C. Signal Detect Algorithm
D. Detailed Steps for Reconfiguring from Mission Mode to Channel Protection Mode
E. Detailed Steps for Reconfiguring from Channel Protection Mode to Mission Mode
F. Hold Timing Violation

C. Signal Detect Algorithm

The signal detect algorithm detailed in this section is an example of how to determine if a valid signal is present at the receiver and then use adaptation to tune the link for acceptable performance. This is only an example and the implementation specifics may need to be modified for your system.

The physical channels to be supported are CEI-28 VSR, a CAUI-4 chip-to-module connection and CEI-56 VSR (PAM4). Other physical channels can be used depending on the application.

The receiver channel must autonomously adapt and lock to the incoming signal when a valid signal becomes present. “Autonomously” means without assistance or information from downstream processing or upper protocol layers, in other words, based exclusively on information available to and via the transceiver itself.

An invalid signal is any situation where the incoming signal is not valid, typically:

  • Module unplugged
  • Fiber unplugged on module
  • No signal received on fiber
  • Bad signal received on fiber

The receiver adaptation completes without disrupting the transmitter path, and a PMA analog reset is not used because PMA analog resets disrupt the transmitter.

The adaptation results in an acceptable BER, for example:

  • CEI-28 VSR: BER 1x10-15
  • CEI-56 VSR: BER 1x10-6

The adaptation completes in:

  • Approximately 100 ms for NRZ
  • Approximately 4 s for PAM4

The entire adaptation process must be able to operate in low power mode. The signal quality is maintained provided that the die temperature variation does not exceed 3 °C per minute and continuous adaptation is enabled.

This process does not deadlock to an unrecoverable state.

Figure 153. Signal Detect Algorithm Flowchart

Valid Signal

For a signal to be valid, both of these conditions must be met:

  • rx_freqlocked_1ms: A filtered version of rx_is_lockedtodata. "Filtered" means that it deasserts instantly on loss of lock and rx_is_lockedtodata needs to be asserted continuously without any deassertions for at least 1 ms before rx_freqlocked_1ms asserts.
  • Eye OK: Requires a read from the transceiver via Avalon® memory-mapped interface and compares a value with an acceptable threshold value of 150 (NRZ) or 25 (PAM4). These units are expressed in steps. The eye threshold values are application-dependent and should be determined by your system validation. See Reading NRZ/PAM4 Eye Height.

If at any point in the algorithm the above valid condition is not met, the adaptation complete status becomes false, and the flow returns to the initial adaptation stage by issuing a one-shot continuous adaptation to stop the continuous adaptation.

Initial Adaptation Stage

The assumption at the start of the test flow is that there may be no signal or an invalid signal may be present. Until an initial adaptation has been run, there is no way to reliably achieve lock, so the initial adaptation process is repeatedly triggered while waiting for a valid signal as per the above definition.

It is possible that the first time a valid signal is detected, the adaptation may not be optimal, for example, because the signal appeared during the initial adaptation process. Therefore, an additional initial adaptation is run following this first detection.

If the signal is still valid after the second initial adaptation, that part of the test flow is complete. When completed, a notification (signal or interrupt) is provided at the system level and to the downstream datapath.

There are two main operations at this stage:

  • Reading the eye height for an eye OK check (this takes <50 µs)
  • Initiating initial adaptation (this typically takes 680 µs from initiating initial adaption to when initial adaptation starts, but it sometimes takes 3 ms)

As the completion time for the initial adaptation varies, the algorithm runs the loop on a fixed timing, triggering a new initial adaptation every 40 ms (chosen to provide some margin) if the loop exit conditions have not been met.

Rarely, if many transceiver channels are in the initial adaptation loop and many of these take a long time to respond on the bus, there may not be enough time to service all channels within this time. In this case, the loop takes longer to complete then continues as usual.

This gives an average completion time of 100 ms, typically, varying between 80 and 120 ms.

For PAM4, the observed completion time is approximately 1.5 seconds using an effort of 0.5.

Note: Average completion times assume that a signal is present. Without a signal, these average completion times are not applicable, and completion may take much longer.

Ongoing Adaptation Stage

In order to maintain good receiver equalization through temperature variations (within specification), an ongoing, non-traffic affecting adaptation process (continuous adaptation) is run while the signal is valid.

Each continuous adaptation takes approximately one second to complete and is repeated automatically while in a continuous mode.

If the signal is removed while in this mode and multiple continuous adaptation operations are performed in this state, the receiver equalization may gradually enter a bad state which cannot be recovered from by re-running an initial adaptation and reaches a deadlock from which this algorithm cannot recover.

To prevent this, the algorithm monitors the rx_freqlocked_1ms signal and immediately reverts to the initial adaptation stage, which issues an initial adaptation which effectively stops the continuous adaptation. Additionally, the eye OK status is polled every second, and, if no longer meeting the criterion, reverts to the initial adaptation stage.

Therefore, at most, one or two continuous adaptation operations execute on an invalid signal which is not enough to bring the transceiver into a bad state.

Internal Serial Loopback

Internal serial loopback cannot be used in LPM. However, there needs to be the ability to switch between internal serial loopback and non-internal serial loopback and perform appropriate adaptation each time.

Internal serial loopback requires Resetting the RX Equalization after setting the internal serial loopback. This RX equalization reset does not impact the transmitter.

The adaptation algorithm must be exited before configuring internal serial loopback and restarted after disabling internal serial loopback and enabling LPM.


Section Content
Signal Detect Algorithm Revision History

Related Information
Level Two Title