02-27-2020 04:18 AM
Hi
I am craeting a bitfile for VU440 FPGA which has DDR4 MIG controller. Bitfile is successfully generated without timing violation. However I am getting a DRC warning as follows.
I am yet to test the bitfile on board. But I am trying to understand what causes this DRC warning and if it will impact DDR4 MIG functionality.
Please share your suggestion.
Thanks,
Aj.
RTSTAT #1 Warning 168 net(s) have no routable loads. The problem bus(es) and/or net(s) are /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_riu/LMB_CE_riu, /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_riu/LMB_UE_riu, /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_top/u_ddr_cal/u_xsdb_arbiter/Q[12], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_top/u_ddr_cal/u_xsdb_arbiter/Q[13], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_top/u_ddr_cal/u_xsdb_arbiter/Q[14], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_top/u_ddr_cal/u_xsdb_arbiter/Q[15], xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_fixdly_rdy_upp/SYNC[0].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_phy_rdy_low/SYNC[0].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_phy_rdy_upp/SYNC[0].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_fixdly_rdy_low/SYNC[0].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_fixdly_rdy_upp/SYNC[1].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_phy_rdy_low/SYNC[1].sync_reg[1], xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_fixdly_rdy_low/SYNC[1].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_phy_rdy_upp/SYNC[1].sync_reg[1], /xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_phy2clb_phy_rdy_low/SYNC[2].sync_reg[1] (the first 15 of 168 listed).
02-28-2020 03:18 PM
Hi @ajasan
If the LMB_CE is not connect (n/c), then the warnig would make sense.
The code snippet shows that it is driven, but does it have loads, or connect to any leaf primitive? I would check this from the open implemented design.
02-27-2020 04:29 AM
Is the IO of the MIG mapped to FPGA pins? Does the IO of the MIG go through the top level of the design to the pads of the FPGA or is the MIG just instantiated and its IO just hang there?
02-27-2020 04:53 PM
Hi @ajasan
The best way to find out what is happening is to trace some specific logic mentioned in the message at each stage of the Build flow.
Does the specific net have a load after synthesis when opening the design?
https://www.xilinx.com/support/answers/58616.html
02-27-2020 09:12 PM
Thanks for the response @bruce_karaffa . Yes, pins are mapped to IO. They are not hanging.
02-27-2020 09:17 PM - edited 02-27-2020 09:31 PM
Thanks for the response @marcb .
This particular net on the message is shown n/c in Implemented design. I checked the RTL. It is driven 1'b0.
This is part of code inside DDR4 SDRAM MIG IP block. Please check the hierarchy as below, if it can give any hint.
$TOP/xilinx_ddr_if_inst/xilinx_ddr4_inst/inst/u_ddr4_mem_intfc/u_ddr_cal_riu
assign LMB_CE = 1'b0; (* dont_touch = "true" *) output LMB_CE
Thanks,
Aj.
02-28-2020 03:18 PM
Hi @ajasan
If the LMB_CE is not connect (n/c), then the warnig would make sense.
The code snippet shows that it is driven, but does it have loads, or connect to any leaf primitive? I would check this from the open implemented design.
03-05-2020 03:07 AM
Its an undriven load and it works fine on board. So I guess the Warning is safe to ignore.
Thank you for the response!
Thanks,
Aj.