ADE XL simluation stuck

Dear Holz,

Holz said:

I'm running a simulation in ADE XL with 3584 point sweeps, it is stuck somehow at 3583....This happens sometimes. After I need rerun the simulation sometimes it doesn't stuck anymore. But for large simulation it's really waste of time.

Are you familiar with the "Re-run Unfinished/Errored points" option available by right-clicking on the results database (in your case Interactive.2)? This does not re-run any of the simulations that completed successfully and only re-runs those corners that did not complete successfully. It copies over the completed results to a new data base win, in your case, a name of something like  "Interactive.2.rerun.0". I just outlined how this is done in response to the following forum post:

community.cadence.com/.../adexl-corner-simulations

Holz said:

Does anybody know the reason of this? I put the screen shot of it and output log of it here.

Speaking for myself, your information is not sufficient for me to guess as to the root cause of your simulation anomaly. A few comments and questions that might help provide some thoughts...

1. As you may know, ADE-XL is no longer supported by Cadence. You might consider migrating to Assembler/Explorer. Assembler is very. very similar and can directly import the ADE-XL state file based onm ADE-L to a new Assember/Explorer maestro view.

2. Are you submitting your jobs using ICRP or LSCS?

3. What are your timeouts in the Job Setup menu set to?

4. Was the netlist for the last job created correctly (i.e., input.scs) that apparently did not complete?

5. Did spectre start for the last job that is apparently stuck? If so, please examine the spectre.out file for errors.

6. Are you sure you did not run out of disk space in the location where you are saving the results? Are you aware there is now a means of checking the available disk space and, should the minimum space you specify not be available, pause your simulations pending an increase in disk space?

Shawn

Some other small suggestions (in addition to what Shawn said):

1. Is the simulation actually still running? Look at the spectre.out (right mouse button menu over the point marked as "running" in the results tab). This unfortunately happens sometimes with ICRP (particularly with larger numbers of jobs) and is one of the reasons for LSCS to be created (note, LSCS is only available in the job options if using Explorer or Assembler, so you'd need to move to that and also best to move to the latest IC618/ICADVM201 hot fix as it was only introduced in ISR9)
2. You might also be able to fix it (although any "running" simulation will re-run if you do this) by going over the little terminal icon in the run summary assistant that is "stuck" and doing right mouse->Stop and Resubmit

Andrew

Some other small suggestions (in addition to what Shawn said):

1. Is the simulation actually still running? Look at the spectre.out (right mouse button menu over the point marked as "running" in the results tab). This unfortunately happens sometimes with ICRP (particularly with larger numbers of jobs) and is one of the reasons for LSCS to be created (note, LSCS is only available in the job options if using Explorer or Assembler, so you'd need to move to that and also best to move to the latest IC618/ICADVM201 hot fix as it was only introduced in ISR9)
2. You might also be able to fix it (although any "running" simulation will re-run if you do this) by going over the little terminal icon in the run summary assistant that is "stuck" and doing right mouse->Stop and Resubmit

Andrew

I just stopped it and did as Shawn said(rerun errored point) and the last simulation was not stuck anymore. Thanks for help