F-Tile Dynamic Reconfiguration Suite Intel® FPGA IP User Guide

ID 711009
Date 9/26/2022
Public

A newer version of this document is available.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

Document Table of Contents

1.2. Design Considerations

  • A reconfiguration is only available within a specific topology. Reconfiguration across different topologies is not supported.
  • A reconfiguration is dependent on the fracturing rules specified in the F-Tile Architecture and PMA and FEC Direct PHY Intel® FPGA IP User Guide. For instance, you can separate a 200G fracture into two independent 100G fractures, and so on.
  • All switching must be done through a neutral state by asserting the digital data path reset and disabling the PMA.
  • Any configuration supports the SerDes rate reconfiguration as long as the rate is legal for a given serialization factor and consistent with any used system clock.
  • Dynamic reconfiguration for wireless IPs applies to TX and RX data paths in a symmetric manner.
  • PMA-direct supports equal PMA widths between TX and RX data paths. However, CPRI supports a PMA-direct width of 20-bit combined with a 32-bit PCS-direct width or 32-bit FEC+PCS-direct width.
  • All IPs supporting dynamic reconfiguration must adapt to the F-tile system clock. You cannot dynamically reconfigure the F-tile system PLLs, including reference clock pin and frequency.
  • The use of Debug Toolkits (NPDME, ETK, and TTK) may prevent successful Dynamic reconfiguration. Intel advises against the use of these Toolkits with protocol IP which is to be dynamically reconfigured.