Stats

  • Locked Locked
  • Replies 2
  • Subscribers 125
  • Views 4211
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

*Error* Library "my_lib" does not exist

Michele Ancis

Hi,

this is the message I get when launching an ADE XL generated OCEAN script from a certain location.

It is the result of a certain procedure outlined below, on which I would like to hear your comments and suggestions.

Thank you!

Background

In some instances I find it easier/faster to perform sweeps on a netlist instead of using the Schematic Editor and the GUI. To accomplish this, if follow these steps:

  1. Create an ADE XL state with a certain test bench, corners and sweeps, variables and all.
  2. Run a sample sweep
  3. Save the OCEAN file of the sweep.
  4. Copy the netlist folder relative to 2. somewhere in my dir structure where I want to have this ran, together with the OCEAN in 3
  5. Modify the OCEAN script as to point to "netlist/netlist" as the design: design("./netlist/netlist")
  6. Launch ocean in the proper directory, then load the modified script

I've seen it working :-)

Meaning, it has worked for me already in the past.

The beginning of the OCEAN script reads as follows:

oceansetXLmode()

ocnxlProjectDir(".")

ocnxlTargetCellView("my_lib" "my_cell" "my_adexl_view")

This should tell the script to use whichever directory it finds itself in as a starting point, and when it works it creates a folder structure from that point with "my_lib" as the top folder, "my_cell" as a sub-folder and "my_adexl_view" as sub-sub-folder with the results structure inside.

Sometimes however I get the error in the title, namely that this library I am asking to use does not exist. One workaround I've found is to specify a different ADE XL view for each time I launch the script: "my_adexl_view1", "my_adexl_view2"...and so on.

It appears as if I cannot overwrite an existing view with new results, or append them.

Is that the intended behaviour? Am I doing something unorthodox?

Cheers,

Michele

  • Andrew Beckett

    Hi Michele,

    When you run in OCEAN XL, the normal expectation is that the target cellView is an existing adexl view. If you do this, it will write a new history into that cellView. Note that for this to work, you can't have ADE XL open in edit mode at the same time on the same cellView because you can't have two sessions at once on the same cellView. So this may be why it's not working for you?

    You can't just change the lib/cell/view arbitrarily - the library name has to exist. If you change the lib/cell/view to a cellView that doesn't exist (in the target cellview call), for a library that does exist, then it won't be completely successful - the situation will run, but you won't be able to open the history. In that case you have to add something like:

    ocnxlSaveSetupAs("opamp090" "NEW_full_diff_opamp" "adexl" )

    (where the lib/cell/view match your target cellView) to save the setup into the new view too.

    Hopefully that will steer you on the right way?

    Regards,

    Andrew

  • Michele Ancis

    EDIT:


    Thanks Andrew,

    yes indeed thanks to your tip I can run flawlessly now.

    I kept experiencing some odd behaviour as reported below, but the reason was that - in the directory created to launch the script - there was no cds.lib file so the system was picking it up...from somewhere not clear (to me), and things just did not work steadily.

    Creating a simple cds.lib with a single #INCLUDE pointing to the right file solved everything.

    Thanks!

    Michele

    ---

    Old Message - Just in case somebody stumbles on a similar situation

    Thanks for the tips Andrew,

    OK I got the point and modified the target CellView to an exsisting adexl view.

    It worked....once!

    [follows a set of desperate attemps at grasping this...]

    It kinda works now, but I experience a funny behaviour I would like to understand better:

    if - to save me time - I try to launch

    > ocean -replay myocn.ocn

    then I keep getting the same old error: Lib not existing.

    If, on the other hand, I start ocean and then load("myocn.ocn") from there, then it works!!

    Why?

    On a second note

    after going past the opening phase, the script indefinitely hangs and never spawns the sim jobs...any high-level thoughts on why that could be? I am of course involving my CAD in this but I know this worked w/o issues only few months ago...

    ...And a third one :-)

    the script launch generates a very large bunch of warnings and - when it goes past opening the CellView - I could see some *error* lines in the screen output. Issue is, there's too many output lines and they get lost due to finite capture of the terminal window.

    I have tried to use ocean -log mylog.log but this file does not contain the information shown during ocean execution, but some higher level summary of what has been going on.

    Any suggestion on how to capture that output? I would like to get to a much cleaner execution.

    Thanks,

    Michele

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.