L-Tile and H-Tile Avalon® Memory-mapped Intel® FPGA IP for PCI Express* User Guide

ID 683667
Date 4/03/2023
Public

A newer version of this document is available.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

Document Table of Contents

9.2. Sending a TLP

The Application Layer performs the following sequence of Avalon® -MM accesses to the CRA slave port to send a TLP Request:
  1. Write the first 32 bits of the TX TLP to RP_TX_REG at address 0x2000.
  2. Set RP_RP_TX_CNTRL[2:0] to 3’b001 to push the first dword of the TLP of the non-posted request into the Root Port TX FIFO.
  3. Write the next 32bits of the TX TLP to RP_TX_REG at address 0x2000.
  4. Set RP_RP_TX_CNTRL[2:0] to 3’b010 if the TPL is completed. Otherwise, set RP_RP_TX_CNTRL[2:0] to 3’b000 to push the next data to the TX FIFO and continue.
  5. Repeat Steps 3 and 4.
  6. When the TLP is completed, the Avalon® -MM bridge will construct the TLP and send it downstream.