Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] LVS weird problem in Cadence

Status
Not open for further replies.

bio_man

Full Member level 2
Joined
Mar 30, 2010
Messages
144
Helped
2
Reputation
4
Reaction score
4
Trophy points
1,298
Activity points
2,698
Hi,

I have built my subcircuits and tested both schematic/layout and LVS and all are ok. However, when I combine them to make a controller bu adding them as a new component/layout, LVS always failed. I removed all pins/labels from subcircuit, LVS still fails. But, when I copied the layout of the subcircuit to the controller layout with pins removed, the LVS works and match!!! what is the main issue here? I couldn't understand what is the problem? The output file keeps reporting cross-mismatch between transistors, but with the same layout copied, this is disappears and LVS match!!
 

Re: LVS wired problem in Cadence,

Sometimes circuits with a lot of identical looking "stuff" that
is repeated, will make LVS confused. This can be totally "luck
of the draw", all about how it recognizes sections as it goes
about netlisting and such.

A layout that is the same at the polygon level, but just
different in some way, could pass where another fails. That's
just the breaks. A layout with pins and a layout without
pins, falls in this situation. LVS is left to decide what the
pin connections are, and it's OK with that - but not OK
with pins asserted?

Seems to me that maybe there's some pin issue or the
schematic matches the "subcircuit" but not the "controller"
(maybe pins, signal nets, power globals, ???). Maybe a pin
ended up misplaced and named some net wrongly.

You can fix it by asserting "correspondence points" (if the
layout and schematic do indeed match) but that effort is
a real nuisance.
 
Re: LVS wired problem in Cadence,

Attached snap shots of the output file, I can't see any issue, the only thing is Net /116 is merged with /42 which I guess not an error. Any hint?
 

Attachments

  • LVS_fails1.png
    LVS_fails1.png
    25.5 KB · Views: 173
  • LVS_fails2.png
    LVS_fails2.png
    18 KB · Views: 183

Re: LVS wired problem in Cadence,

"Merged" usually means that LVS thinks these two nets
should be one. You should go to the extracted view,
select-by-property the two nets and see if maybe
there is a wiring "miss" (gap, or pins that should
connect are missing the wire to do so altogether).
 
Re: LVS wired problem in Cadence,

What seems an issue to me is a warning about "pcapacitor" being in the netlist but not in the layout.
"pcapacitor" and "presistor" elements are supposed to be eliminated during netlisting step - but this should be configured properly in your environment.
These elements are added to schematic to account for parasitic elements (and for net naming purposes - where you can have several names for the same net, by inserting artificial resistors), prior to parasitic extraction.
 
Thanks all. I solved this issue by making another layout and copy some cells (which make LVS fails) without their pins, It works now.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top