F-Tile Ethernet Intel® FPGA Hard IP User Guide

ID 683023
Date 12/19/2022
Public

A newer version of this document is available.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

Document Table of Contents

4.4.3.1. PTP TX Client Flow

In this section, the acronyms PL and VL stand for Physical Lane and Virtual Lane respectively.

The following flows depict pseudo-code meant for the conceptual, illustrative purposes. For definitive software routines, refer to the design example.

Important: If IP undergoes TX reset at any point in this flow, you must restart the entire PTP TX client flow.
  1. After power up or a TX reset, wait until TX raw offset data are ready.
    You can monitor the status via one of the following:
    • Output port:
      o_tx_ptp_offset_data_valid = 1'b1
    • Polling via Avalon® memory-mapped interface register until it is asserted:
      csr_read(ptp_status.tx_ptp_offset_data_valid) = 1’b1
  2. Read TX raw offset data from IP:
    tx_const_delay      = csr_read(ptp_tx_lane_calc_data_constdelay[30:0])
    tx_const_delay_sign = csr_read(ptp_tx_lane_calc_constdelay[31])
    
    for (pl = 0; pl < PL; pl++) {
     tx_apulse_offset[pl] = csr_read(ptp_tx_lane<pl>_calc_data_offset[30:0])
     tx_apulse_offset_sign[pl] = csr_read(ptp_tx_lane<pl>_calc_data_offset[31])
     tx_apulse_wdelay[pl] = csr_read(ptp_tx_lane<pl>_calc_data_wiredelay[19:0])
     tx_apulse_time[pl]   = csr_read(ptp_tx_lane<pl>_calc_data_time[27:0])
    }
  3. Determine TX reference lane:

    The following sub-steps apply for designs with multiple lanes since any lane can be used as reference lane. You can skip the sub-steps for designs with one single PMA lane by setting the tx_ref_pl = 0.

    1. Detect rollover of asynchronous pulse time:

      The tx_apulse_time[pl] signal represents an asynchronous time of each physical lane in a 28-bit format, where bit [27:16] represent asynchronous pulse time in nanoseconds (ns) and bit [15:0] represent asynchronous pulse time in fractional nanoseconds (fns).

      Two types of rollover are possible:
      1. Natural rollover from bit 27 to bit 28 when the value reaches 28'hFFF_FFFF. Before rollover, bit [27:24] is 4'hF.
      2. Billion rollover when the TOD reaches one billion ns or 48'h3B9A_CA00_0000 in hex value. Before rollover, bit [27:24] is 4'h9.
      Given tx_apulse_time_max is largest tx_apulse_time from all physical lanes,
      for (pl = 0; pl < PL; pl++){
        if (tx_apulse_time_max - tx_apulse_time[pl] > 29'h01F4_0000){
          tx_apulse_time[pl] = tx_apulse_time[pl] + 29'h1000_0000
        } else {
          tx_apulse_time[pl] = tx_apulse_time[pl] + 29'h0A00_0000
        }
      }
    2. Calculate the actual time of TX Alignment Marker at TX PMA parallel data interface.
      for (pl = 0; pl < PL; pl++) {
       tx_am_actual_time[pl] = 
         (tx_apulse_time[pl]) 
       + (tx_apulse_offset_sign[pl] ? –tx_apulse_offset[pl] 
                                     : tx_apulse_offset[pl])
       – (tx_apulse_wdelay[pl])
      }
      
    3. Determine TX reference lane:
      tx_ref_pl = pl 
      The TX reference lane is the TX physical lane containing the largest tx_am_actual_time when comparing among all physical lanes.
  4. Calculate TX offsets:
    Attention: Step 4c is not applicable for 10G and 25G Ethernet data rates. You must skip step 4c for these rates.
    1. Calculate TX TAM adjust:
      tx_tam_adjust_sim = 
        (tx_const_delay_sign ? –tx_const_delay : tx_const_delay)
      + (tx_apulse_offset_sign[tx_ref_pl] ? 
            –tx_apulse_offset[tx_ref_pl] : tx_apulse_offset[tx_ref_pl])
      – (tx_apulse_wdelay[tx_ref_pl])
      
      For hardware run with PTP Timestamp accuracy mode set to Advanced:
      tx_tam_adjust = (tx_tam_adjust_sim) 
      + (tx_routing_adj_sign[tx_ref_pl] ? – tx_routing_adj[tx_ref_pl] 
                                          : tx_routing_adj[tx_ref_pl])
      For routing delay adjustment information, refer to the Routing Delay Adjustment for Advanced Timestamp Accuracy Mode section.
      For all other cases:
      tx_tam_adjust = tx_tam_adjust_sim 

      Convert TAM adjust to a 32-bit 2's complement number:

      tx_tam_adjust_2c = tx_tam_adjust
      where tx_tam_adjust is a 32-bit 2's complement number
    2. Calculate TX extra latency:
      Convert unit of TX PMA delay from UI to nanoseconds. For UI value, refer to tables specified in UI Value and PMA Delay.
      tx_pma_delay_ns = tx_pma_delay_ui * UI13
      TX extra latency is a positive adjustment. To indicate the positive adjustment, set the most-significant register bit to 0. Total up all extra latency together:
      tx_extra_latency[31] = 0
      tx_extra_latency[30:0] = tx_pma_delay_ns + tx_external_phy_delay
    3. Calculate TX virtual lane offsets:

      Use VL0 as the reference virtual lane. Assign TX virtual lane offset values according to virtual lane order.

      • For KP-FEC or LL-FEC variants:
        Note: % is the modulo operator.
        
        for (vl = 0; vl < VL; vl++) {
           tx_vl_offset[vl] = [vl - (vl % PL)] / PL * 68 * UI13
        }
      • For KR-FEC variants:
        for (vl = 0; vl < VL; vl++) {
           tx_vl_offset[vl] = [vl - (vl % PL)] / PL * 66 * UI13
        }
      • For no FEC variants:
        for (vl = 0; vl < VL; vl++) {
           tx_vl_offset[vl] = [vl - (vl % PL)] / PL * 1 * UI13
        }
  5. Write the determined TX reference lane into IP:
    csr_write (ptp_ref_lane.tx_ref_lane, tx_ref_pl)
  6. Write the calculated TX offsets to IP:
    Attention: Step 6a is not applicable for 10G and 25G Ethernet data rates. You must skip step 6a for these rates.
    1. Write TX virtual lane offsets:
      for (vl = 0; vl < VL; vl++) {
         csr_write(tx_ptp_vl_offset_<vl>, tx_vl_offset[vl])
      }
    2. Write TX extra latency:
      csr_write(tx_ptp_extra_latency, tx_extra_latency)
    3. Write TX TAM adjust:
      csr_write(ptp_tx_tam_adjust, tx_tam_adjust_2c)
  7. Notify soft PTP that uses flow configuration is completed.
    csr_write(ptp_tx_user_cfg_status.tx_user_cfg_done, 1'b1)
  8. UI value measurement. Follow the steps mentioned in the TX UI Adjustment section..

    For simulation or hardware run with 0 ppm setup, you can skip the measurement and program 0 ppm UI value defined in UI Adjustment.

  9. Wait until TX PTP is ready.
    You can monitor the status via one of the following:
    • Output port:
      o_tx_ptp_ready = 1'b1
    • Polling via CSR:
      csr_read(ptp_status.tx_ptp_ready) = 1’b1
  10. TX PTP is up and running.
    1. Adjust TX UI value.

      Perform the TX UI adjustment occasionally to prevent time counter drift from golden time-of-day in the system. Follow the steps described in TX UI Adjustment .

      Note: UI measurement is a long process in simulation. Therefore, for simulation, Intel recommends skipping this step and program a 0 ppm value.
13 The UI format differs from the format of other variables. UI uses the {4-bit ns, 28-bit fractional ns} format. Other variables defined in this flow use the {N-bit ns, 16-bit fractional ns} format, where N is the largest number to store the calculation's max value. If you use UI format in your calculation, you must convert your result to a 16-bit fractional ns format.