It looks like you're new here. If you want to get involved, click one of these buttons!
Hi Matthias,
By using following sample data, I confirmed LVS normal operation.
https://github.com/KLayout/klayout/tree/master/samples/lvs.
Then, I added dummy poly data at both virtical side of INVX1 cell.
This data insertion is common in advanced processes.
As a result, extra pins are generated and the circuit verification
of ringo circuit cannot be completed.
.SUBCKT INVX1 VDD OUT VSS \$4 nc_1 nc_2 IN SUB
M$1 VDD IN OUT \$4 PMOS L=0.25U W=1.5U AS=0.6375P AD=0.6375P PS=3.85U PD=3.85U
M$2 VSS IN OUT SUB NMOS L=0.25U W=0.95U AS=0.40375P AD=0.40375P PS=2.75U PD=2.75U
.ENDS INVX1
How can I suppress this pin generation?
Thanks,
Comments
Hi haru_f,
these floating nodes should be eliminated from the netlist if you use "netlist.simplify" before the compare step.
Such dummy pins are very common not only in dummy gates but also in filler cells or dummy pattern for uniform density.
Have you included this step?
Matthias
Hi Matthias,
Thank you for your comment.
Yes, "netlist.simply" is used. I use the template script which Klayout provided.
I think that since dummly poly data of the INVX1 cell is completely overlapped with that of the adjacent cell in this case, the terminal is generated to connect
each terminal by poly.
Sorry for typo.
*Netlist normalization
netlist.simplify
Thanks,
haru_f
This is the result of LVS.
"$4", "$5" pins are remain in "layout" netlist.
Thanks,
Another trial result:
Dummy poly data is added at both side of ND2X1.
Since the right side dummy poly is connected to the adjacent INVX1's dummy poly, "$4" pin is generated in ND2X1
while the left side data has been removed as a floating node.
Thanks,
haru_f
I think I see the issue. I need to look into this. I'd say it's a bug related to "simplify". The case of floating nets connected together apparently isn't captured yet.
I have created an issue for this: https://github.com/KLayout/klayout/issues/425
Kind regards,
Matthias
Hi Matthias,
I hope it will be improved.
Best regards,
haru_f
There is a fix for this issue. It's a deep modification however and I hope it does not break anything else. See the GitHub link for details.
Matthias
Thank you!
haru_f
Hi Matthias,
I could confirm this issue (Enhancement: #425) with my example data.
Thank you very much for your effort.
Best regards,
haru_f