EMIR simulation fails with DSPF-114 code

Hi,

As the error message suggests, did you check if there is an Instance section in the DSPF? Instance section contains all the devices (transistors, design resistors, design capacitors etc) and their connectivity to the nets in the design. If this section is missing, it means that your extraction setup isn't correct. For EMIR flow, you can find the required Quantus/QRC settings in the Rapid Adoption Kit for Voltus-Fi-XL.

spfchecker looks through the first subckt definition in the DSPF which is empty in your case. Please run it again as: spfchecker -ckt top_level

Regards,

Saloni

Hi Saloni,

Thanks for the reply.

I checked the DSPF file, and there are 2 lines that have:

* Instance section

The first time this occurs is within the dig_block subckt section, and it looks like:

* instance section

*

.ENDS dig_block

The second occurrence is within the top_level subckt section, and it looks like:

* instance section

*

X....etc (many lines of what looks like transistor information)

So is the problem that, as you pointed out in the last line, that the spfchecker and EMIR analysis is only looking at the first subckt, which is for the dig_block, and is empty of extracted info, and which has an empty instance section ?

If that's the case, should I use option -ckt top_level in the highlighted field of the EMIR setup (in the attached pic)?

Thank you very much

Yes, you can use -ckt argument in the field you highlighted, and that will tell spfchecker which subckt to check. However, that still does not mean that EMIR analysis will work.

You also need to add black_box="dig_top" argument in dspf_include statement in Spectre netlist. Are you including the DSPF through Simulation Files in ADE? If yes, then you can provide the black-box name as shown in snapshot below (make sure you hit apply so it becomes visible under Options column). You can also check dspf_include statement in input.scs to make sure that you've got argument there as: dspf_include "/path/to/top.dspf"   black_box="dig_top"

Thanks a lot Saloni, that did the trick.

Now I'm getting another error:

Error found by spectre during circuit read-in.
    ERROR (SFE-461): "/Desktop/dig_block.scs" 7: Port number of subckt `dig_block' mismatch between schematic netlist and spf file. Port number in schematic netlist is 1249, and in spf file is 3273.

Should I create a different topic for this?

Thank you

Thanks a lot Saloni, that did the trick.

Now I'm getting another error:

Error found by spectre during circuit read-in.
    ERROR (SFE-461): "/Desktop/dig_block.scs" 7: Port number of subckt `dig_block' mismatch between schematic netlist and spf file. Port number in schematic netlist is 1249, and in spf file is 3273.

Should I create a different topic for this?

Thank you

The error message says that the number of ports found in Spectre netlist for 'dig_block' is much smaller than the ports found in the empty subckt definition in DSPF. This can happen when you run black-box extraction. To get past the error, you can try setting option 'extra_port=true'. This is available under DSPF Options in Simulation Files setup (just a row above where you specified Black Box name). Any extra ports in DSPF will be ignored and will not be driven by the testbench/design.

That's the best I can advise without looking at the design and setup. Your simulation results should tell you if the port mismatch is causing any issues or not.

Regards,
Saloni